KunPeng: A Global Ocean Environmental Model

2025-04-07

AI Summary

This paper addresses three key challenges in global ocean environmental forecasting: difficulty in modeling spatial discontinuities, insufficient coupling of multi-scale spatiotemporal features, and weak representation of temporal dependencies. To this end, we propose a physics-informed deep forecasting framework. Our method introduces a topography-adaptive masking constraint, integrates a longitude-cyclic deformable convolutional network (LC-DCN), and designs a deformable-convolution-enhanced multi-step prediction module (DC-MTP). It further incorporates meteorological large-model transfer learning, dynamic receptive fields, and a multi-scale feature pyramid. Evaluated on 15-day global sea surface temperature, salinity, and current velocity field forecasting, the framework achieves an average anomaly correlation coefficient (ACC) of 0.80, with MSE and MAE reduced by 5–31% and 0.6–21%, respectively. Notably, it significantly improves prediction accuracy in deep-ocean and complex-topography regions, establishing a novel paradigm for high-resolution global ocean physical modeling.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Deep Learning Theory
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Inspired by the similarity of the atmosphere-ocean physical coupling mechanism, this study innovatively migrates meteorological large-model techniques to the ocean domain, constructing the KunPeng global ocean environmental prediction model. Aimed at the discontinuous characteristics of marine space, we propose a terrain-adaptive mask constraint mechanism to mitigate effectively training divergence caused by abrupt gradients at land-sea boundaries. To fully integrate far-, medium-, and close-range marine features, a longitude-cyclic deformable convolution network (LC-DCN) is employed to enhance the dynamic receptive field, achieving refined modeling of multi-scale oceanic characteristics. A Deformable Convolution-enhanced Multi-Step Prediction module (DC-MTP) is employed to strengthen temporal dependency feature extraction capabilities. Experimental results demonstrate that this model achieves an average ACC of 0.80 in 15-day global predictions at 0.25$^circ$ resolution, outperforming comparative models by 0.01-0.08. The average mean squared error (MSE) is 0.41 (representing a 5%-31% reduction) and the average mean absolute error (MAE) is 0.44 (0.6%-21% reduction) compared to other models. Significant improvements are particularly observed in sea surface parameter prediction, deep-sea region characterization, and current velocity field forecasting. Through a horizontal comparison of the applicability of operators at different scales in the marine domain, this study reveals that local operators significantly outperform global operators under slow-varying oceanic processes, demonstrating the effectiveness of dynamic feature pyramid representations in predicting marine physical parameters.

Problem

• Develops ocean prediction model using atmospheric techniques
• Addresses training divergence at land-sea boundaries
• Enhances multi-scale oceanic feature modeling

Innovation

• Terrain-adaptive mask for land-sea boundaries
• Longitude-cyclic deformable convolution network
• Deformable Convolution-enhanced Multi-Step Prediction